T
We've just updated Vision Vanguard 25 (ERNIE Image) to V2026.2. Trained using a new methodology, we trained the LoRA at a lower LR. This prevented overfitting - removing artifacts as well as errors. Try it out and let us know what you think!
Top safe models by thedivergentai on CivitAI
Explicit content is hidden · show all
Showing 1-1 of 1 results
